A place for developers to talk about building on Farcaster.

/dev

Devs are waking up 🪖

https://youtu.be/zJVyOETKvzQ

DX Engineer | DevRel at OpenZeppelin | Building orbiter.host | stevedylan.dev/links | opinions are my own

$20/mo to not have to do any of this is a great deal

🔍 ensdata.net ✦ 𝙽𝙾𝚆: design engineer @family ✦ 𝙿𝚁𝙴𝚅: @rainbow 🌈 ✦ 𝙱𝚄𝙸𝙻𝙳𝙸𝙽𝙶: @vision

Until it's not $20 a month ;) There are plenty of other solutions out there that work way better than this and are still very affordable (i.e. Cloudflare workers or pages). The key imo is using frameworks that have the ability to migrate somewhere else vs Next which is a total lock in

i ran into bugs with some middleware configurations when hosting next.js outside of vercel but managed to work around them. everything else was working properly, no rendering bugs. 

but yeah vercel + next is a great bundle. they are making a lot of things so easy — especially for personal software or apps that won’t have 10k+ users. 

what options are those? 

using vite with react makes the most sense for highly dynamic apps. but you can achieve the same results with next.js although you need to know the framework and platform to prevent paper cuts which can end up costing $$$

While you can deploy Next in other contexts, it's never complete and has the full integration that Vercel has with Next. Sure you can limit SSR as much as possible, but you still end up with a messy experience full of bugs (which people do, and hence why OpenNext exists). Why bother when there are way better options out there? I totally understand the appeal of Next.js and Vercel, but imo they're a bundled deal.

next can be deployed to railway at any time. the key is decoupling your backend from serverless functions and using SSR only where it makes sense